problem at maintaining geometry coordinate

From: Kwee Hong (joyssstan0202_at_gmail.com)
Date: Mon Aug 23 2010 - 02:22:53 CDT

Hi all,

I'm trying to simulate a metallothionein which is bound to 3 cadmium ion
with its 9 cysteines. Each cadmium is coordinated tetrahedrally by 4
cysteines. As the cysteines in the protein are deprotonated and bound to
either one or two cadmium, adjustment had been made in the RESI CYS to
yeildRESI CYT (terminal thiol) and RESI CYB (bridging
thiol)<http://birg1.fbb.utm.my/joyce/modifiedcys.html>.
A patch <http://birg1.fbb.utm.my/joyce/patch.html> had been added to the
topology so that the .pdb and .psf can be produced to initiate the MD
simulation. After generating the necessary psf and pdb file using
script<http://birg1.fbb.utm.my/joyce/genpsf.html>,
the system was then solvated and minimized for 500 steps at 0 temperature
while pressure was allowed to fluctuate. Before the minimization can be run
successfully, I've done some adjustment to the parameter file (
par_all22_prot.inp <http://birg1.fbb.utm.my/joyce/modpar.html>).Then, the
system is subjected to heating for 6ps where the temperature increased 20K
every 100 steps until it reaches 300K. Then the system is equilibrated for
20ps before the MD run for 1ns. At the end of the simulation,the Cd-S bond
length had been overestimated by the force field and the metal-thiolate
center has lost its tetrahedral coordination. The Cd-S bond length has
become more than 0.6nm. According to experiment, the bond length and bond
angle should be as follow:

Cd-S(terminal) 0.2467nm
Cd-S(bridging) 0.2560nm

S-CD-S 109.5
Cd-S-C 95.5

I would need some advice on how to keep the geometry coordination at the
metal-thiolate center not to fluctuates too much from the experimental
value. Is there any problem when I was building the patch? Or when I was
modifying the topology or parameter file? Link in this email would bring you
to the relevant script or modification I've made. Log file of 250ps can be
reached through this link <http://birg1.fbb.utm.my/joyce/mdlog.html>.

Any advice are welcomed and appreciated.

Thanks.

Regards,
Joyce

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:26 CST